Projects

Open source projects and research platforms.

OpenR

⭐ 1,836

Open-source framework for advanced LLM reasoning via process reward models, tree search, and reinforcement learning.

LLMReasoningRL

SMARTS

⭐ 1,112

πŸ† Best System Paper, CoRL 2020

Scalable multi-agent RL training school for autonomous driving. Powered NeurIPS 2022 driving challenge.

Autonomous DrivingSimulationMARL

MALib

⭐ 550

JMLR 2023

Parallel framework for population-based multi-agent RL. Supports self-play, league training, and PSRO.

MARLFrameworkPopulation-based

MAT (Multi-Agent Transformer)

⭐ 489

NeurIPS 2022

Casting multi-agent RL as a sequence modeling problem. SOTA on cooperative benchmarks. Deployed in real-world multi-AGV warehouse systems.

TransformerMARLDeployed

ZSC-Eval

⭐ 55

NeurIPS 2024

Evaluation toolkit and benchmark for multi-agent zero-shot coordination.

EvaluationZero-ShotMARL

GEAR

⭐ 19

ICML 2023

GPU-centric experience replay system for large RL models. 6Γ— throughput over DeepMind Reverb.

SystemsGPURL

Voice Memo Sync

⭐ 6

Intelligently sync, transcribe, and organize Apple Voice Memos with AI. OpenClaw Skill.

AIVoiceProductivity